Space is by far the most dangerous working environment known. In early 2013, more than 530 people have travelled to space, spending a total of 29,000 person-days there. 18 have died in space accidents (mostly during launch or re-entry) and 11 have died in training. Another 28 have died on the ground preparing the missions, and hundreds have died developing rocket technology.
This does not include the thousands who died in the Nordhausen slave labor camps building the V2 rocket, or the thousands who died when those weapons hit their targets. The death toll would have been even far ghastly had those same slaves died producing more accurate and effective weapons. Nor does it include the speculative billions who would have died in a nuclear ICBM exchange between the US and the Soviet Union, or who may die if madmen launch nuclear weapons in the future. This short essay is about the hazards of space travel in peacetime.
If we consider only the risk to space travellers themselves, and ignore the lingering health effects of zero gee and radiation exposure, the result is one death per 1000 days in space, or about 27% per year. This is a huge number. If we include the people who die on the ground to make these missions possible, the death rate is 54% per year. The highest death rate for any other vocation is deep ocean fishing, with a death rate of 0.12%. Space travel is 300 times more dangerous per year in orbit.
I have always dreamed of living and working in space, but the cold reality is that space is more dangerous than the front line trenches in a major war. The underlying reason is clear: the energy and power levels associated with entering and leaving orbit are enormous, and that energy can turn into a huge explosion very quickly with the failure of a minor component. 29,000 person days in orbit is 80 person-years. The sum of all time ever spent in orbit is about the same as one human life. That is a vanishingly small percentage of all human experience.
In time, we will multiply that experience, both in duration and in death. Hopefully, duration will increase far faster. However, almost all deaths have been during launch and reentry, or during launch prep. If "space tourism" results in 1% of the current death rate per flight, and the time spent in "space" is a few minutes per flight, we can expect the death rate to increase above 200 per person-year in orbit. This sounds less like a lifetime dream and more like the life expectancy of a kamikaze pilot.
A resourceful human can live without tools or clothing on a desert island for years - in space, death happens in a fraction of a minute. We should never give up on the dream of living in space, but we should be clear that this is not at all like colonizing a fertile patch of land across an ocean of water. Even for terrestrial colonization, the survival rates were poor.
Robots are cheap. We should launch millions of robots, let most of them die, and learn to travel space safely before we plan to settle space permanently. Launch loops will launch those millions of robots faster and cheaper than rockets, so our wait may be decades rather than centuries if we focus on finding paying markets for those robots.